Mechanisms of action of second-line agents and choice of drugs in combination therapy

E. Choy, G. Panayi

Department of Rheumatology, Division ABSTRACT stimulation of IL-2 receptor (IL-2R) of Medicine, GKT School of Medicine, Second-line agents are used commonly positive lymphocytes and monocytes. King’s College, London. for the treatment of rheumatoid arthritis The latter release monokines, including Dr. Ernest Choy, MD, MRCP, Consultant (RA). They suppress inflammation and IL-1, IL-6, and tumour necrosis factor a and Senior Lecturer in Rheumatology; ameliorate symptoms but often fail to (TNFa ) that stimulate mesenchymal Professor Gabriel Panayi, MD, DSc, substantially improve long-term disease cells such as fibroblasts, as well as en- FRCP, Arthritis Research Campaign outcome. Their use in RA was discov- dothelial cells. Activated fibroblasts, mo- Professor of Rheumatology. ered serendipitously and their modes of nocytes, and macrophages release ma- Please address correspondence and action were largely unknown. Recent re- trix metalloproteinases, such as collagen- reprint requests to: Dr. E. Choy, searches have identified some of their ases and stromelysin, that degrade con- Department of Rheumatology, GKT mechanisms of action. Most of them have nective tissues and result in tissue dam- School of Medicine, King’s College antiinflammatory properties and some age. Stimulated endothelial cells up-re- Hospital (Dulwich), East Dulwich are immunomodulators. Traditionally, gulate surface vascular adhesion mole- Grove, London SE22 8PT, U.K. second-line agents are used as mono- cule expression. These include selectins

Clin Exp Rheumatol 1999; 17 (Suppl. 18): therapy, but recent evidence suggests and integrins such as intercellular adhe-

S20 - S28. that combination treatment with two or sion molecule 1 (ICAM-1) and vascular © Copyright CLINICAL AND more drugs may be more efficacious. cell adhesion molecule 1 (VCAM-1). EXPERIMENTAL RHEUMATOLOGY 1999. However, the choice of agents in combi- This leads to increased recruitment of nation therapy is not based on their leukocytes into the inflammatory site and Part of this article has been previously published in “How do second- line agents mechanisms of action. We review current augments the immune response. work ?” by E.H.S. Choy and G.H. Kingsley, knowledge on the modes of action of sec- The release of chemokines such as IL-8 in Immunology of Rheumatoid Disease, ond-line agents and assess whether such further augments leucocyte trafficking edited by J. Saklatvala and M.J. Walport, understanding may offer a rational ba- into the inflamed joint. IL-6 stimulates Edinburgh, Churchill Livingstone, sis for combination therapy. hepatocytes to release acute phase reac- 1995; 51 (2): 472-492. tants and is the main cytokine responsi- Introduction ble for the raised erythrocyte sedimen- Key words: Second-line or disease-modifying anti- tation rate (ESR) and C-reactive protein Rheumatoid arthritis, treatment, rheumatic drugs (DMARDs) are stand- (CRP) in RA. Within this inflammatory combination, DMARD. ard treatment for rheumatoid arthritis milieu, there are factors, such as trans- (RA). Some, but not all, DMARDs re- forming growth factor b (TGFb) and IL- tard joint damage (1). For all the current 10, which down-regulate inflammation. DMARDs, their use in RA has come The degree of inflammation in RA rests about through serendipity rather than on the balance between the pro-inflam- rational development. This, unfortunate- matory and antiinflammatory mediators. ly, reflects our ignorance of the precise The exact role of B cells in the patho- etiopathogenesis of RA. The exact mech- genesis of RA is unclear, but the pres- anism of action of most DMARDs re- ence of rheumatoid factor (RF) and hy- mains unknown. pergammaglobulinemia in RA patients Recent researches have increased our suggests that they may have a role in the understanding of inflammation and path- inflammatory process. A high titre of RF ogenesis of RA. This has been reviewed is associated especially with a poor prog- extensively elsewhere (2, 3). Briefly, RA nosis, erosive disease, and extra-articu- is thought to be driven by unknown anti- lar manifestations. genic peptides presented in the groove Clinical trials of DMARDs have con- of human leucocyte antigen (HLA)-DR firmed their ability to reduce joint pain, molecules to CD4+ T lymphocytes. swelling, and early morning stiffness, all These antigen-specific CD4+ T cells re- of which are features of inflammation. lease lymphokines such as interleukin-2 Hence, most research efforts have fo- (IL-2) and interferon-g (IFNg) to activate cused on the effects of DMARDs on the inflammatory cascade through the various inflammatory mediators. In or-

S-20 Hypothalamus-pituitary-adrenocortical and -gonadal axis in RA / M.Mechanisms Cutolo of action of second-line agents / E. ChoyEDITORIAL & G. Panayi der to elucidate the mechanism by which tion in the number of monocytes (8). Preincubating monocyte or B cell enrich- a particular DMARD acts in RA, it is Furthermore, in vitro aurothiomalate is ed PBMNC with aurothiomalate inhib- vital to obtain both in vivo and in vitro a potent inhibitor of peripheral blood ited the mitogen stimulated production evidence. On its own, the former may mononuclear cell proliferation induced of immunoglobulin in vitro (18). This is reflect merely disease improvement or by mitogen, antigen, or mixed allogeneic confirmed by in vivo and ex vivo data other factors rather than a direct action lymphocytes (9-12). This is, at least in showing that chrysotherapy reduces im- of the second-line drug. The results of part, a direct effect on monocytes, since munogobulin, immune complexes, and in vitro studies vary greatly with the ex- addition of untreated monocytes could RF levels in serum (19, 20). Further- perimental model used, which may not reverse the inhibition (10). Although more, the side effects of gold include hy- truly reflect the complex in vivo situa- such monocytes showed the decreased pogammaglobulinemia and selective IgE tion. Furthermore, any in vitro effect expression of major histocompatibility and IgA deficiency (21). must be demonstrable at doses that complex (MHC) class II molecules, this The effect of gold compounds on cyto- would be achieved in patients. This may may not be the mechanism of action in kine expression remains unclear. In vivo be difficult to determine for some prepa- vivo since a high dose of aurothiomalate Madhok et al. found that after chryso- rations. These caveats should be borne was required in vitro to produce such therapy there was a reduction in serum in mind in assessing many apparent con- effects (10). IL-6 which was correlated with disease flicts in the current literature. Although aurothiomalate inhibits mito- activity, but this may be due to disease gen-stimulated peripheral blood mono- improvement rather than a direct effect Gold nuclear cell (PBMNC) proliferation and of gold (22). Farahat et al. studied se- On the assumption that RA was caused expression of IL-2 and IL-2R, hitherto quential synovial biopsies and showed by mycobacterial infection, Forestier et there is no convincing direct evidence that there was a significant decrease in al. treated the first RA patient with gold to suggest that aurothiomalate exerts its the expression of IL-1a , IL-1b, IL-6, and in 1929 (4). The clinical efficacy of gold prime therapeutic effect through a direct TNFa after 12 weeks of chrysotherapy in RA was demonstrated by a multicen- action on lymphocytes (12). Interest- (8). In vitro, Harth et al. found that auro- tre trial in the UK published in 1960 (5). ingly, Verwilghen et al. showed that lym- thiomalate inhibited the production of are administered either orally phocytes from patients who developed IFNg from concanavalin A-stimulated as or intramuscularly as sodi- gold-induced skin rashes proliferated peripheral blood mononuclear cells both um aurothiomalate in the UK, and as so- specifically to gold (13). Surprisingly, in RA patients and normal controls (23). dium in some European this response was directed against gold Danis et al. showed that aurothiomalate countries. Whilst intramuscular gold (III) rather than gold (I), which is the had a bimodal effect on IL-1 production compounds appear to have similar mech- form actually found in aurothiomalate by lipopolysaccharide-stimulated mono- anisms of action and pharmacokinetics, (13). It was therefore postulated that gold cytes (24). At low concentrations, the it is unclear to what extent auranofin was oxidized from gold (I) to gold (III) production of IL-1 was enhanced, while shares these characteristics. in vivo, perhaps in the phagolysosomes higher concentrations had the opposite The primary therapeutic effect of gold of monocytes, macrophages, and neutro- effect. These changes in cytokines could, salts appears to be on polymorphonu- phils. Of particular interest, the patients of course, be secondary to the effect of clear cells (PMN), monocytes, and en- who developed skin rashes as a result of aurothiomalate on monocytes. However, dothelial cells, though they may also af- chrysotherapy tended to be those who one intriguing mode of action of gold fect B cells and cytokines. responded particularly well to treatment may be the direct inhibition of DNA In vitro aurothiomalate and auranofin at (14), suggesting that the induction of this binding by transcription factors (TF) pharmacologic doses inhibit PMN pha- gold (III)-specific T cell response might (25). TF such as activator protein 1 (AP- gocytosis, aggregation, chemilumines- also be linked to a therapeutic mode of 1) and, to a lesser extent, AP-2, nuclear cence, and the generation of superoxide action of aurothiomalate. factor 1 (NF-1), and TFIID are involved (6). Interestingly, auranofin is a more Gold is known to affect endothelial cell in the production of cytokines and have potent inhibitor than aurothiomalate in proliferation and HLA expression (15). been shown to be inhibited in vitro by these systems, whilst the latter is more It has also been shown to affect leucocyte gold (25). effective clinically (7), suggesting that trafficking in vivo using the air pouch Gold salts may also inhibit the prolifera- inhibition of PMN function is not a ma- model (16). In support of this, Corkill et tion of cultured synovial cells and the jor therapeutic effect of aurothiomalate. al. showed by synovial immunohistolo- synthesis of collagen in vitro; this may Auranofin inhibits the depolarisation of gy in patients receiving aurothiomalate be especially important in the inhibition stimulated PMN, while aurothiomalate that there was a significant reduction in of pannus formation (26). stimulates the oxidative burst but has no the expression of endothelial leucocyte effect on membrane depolarisation. adhesion molecule 1 (ELAM-1) on high Another major effect of aurothiomalate endothelial venules (17) in the joint Methotrexate (MTX) is an inhibitor of is on monocytes. Synovial biopsies per- which would be expected to reduce cel- folate metabolism that has been used tra- formed before and after aurothiomalate lular trafficking. ditionally primarily in the treatment of therapy showed that there was a reduc- Gold salts may have effects on B cells. malignancies. At high doses, it suppress-

S-21 Mechanisms of action of second-line agents / E. Choy & G. Panayi es proliferation of cells by inhibiting es- serum IL-6, soluble IL-2, and p55 TNF function. sential enzymes such as dihydrofolate re- receptors (38). In collagen-induced ar- A second proposed mechanism of action ductase, thymidylate synthetase, and thritis, MTX reduced spontaneous and of D-pen involves the formation of per- aminoamidazole carboxamide ribonu- IL-15-induced TNFa production by oxides. Some investigators have suggest- cleotide transformylase in the folate splenic T cells and macrophages (39). ed that the oxidation of D-pen in the pres- metabolic pathway (27, 28). However, There is conflicting evidence as to whe- ence of trace copper leads to the devel- at low doses such as those used in the ther MTX can inhibit IL-1b in vitro (40, opment of a reactive oxygen species, treatment of RA and psoriasis, its immu- 41). Seitz et al. showed that MTX in- peroxide, which affects the function of nosuppressive and antiinflammatory ef- duces monocytes to differentiate into T lymphocytes, endothelial cells, and fects are largely unrelated to these en- macrophages in vitro, associated with an fibroblasts (52-54). However, the exact zymes. increase in IL-1Ra and sTNFR p75 re- relevance of this observation remains One of the most significant antiinflam- lease and a decrease in IL-1b (42). Oth- unclear, as the concentration of D-pen matory actions of MTX is the inhibition ers showed that MTX did not affect the used in these experiments was greater of leucocyte trafficking. Following MTX production of IL-1 from murine macro- than that achieved in vivo. treatment, there is a decrease in PMN phages and RA patients’ PBMNC, but D-pen also inhibits leucocyte myeloper- infiltration in the skin and joints of pa- MTX seemed to inhibit IL-1ß activity in oxidases which catalyse the formation of tients with psoriatic arthritis (29) and RA functional assay (43). One possible ex- hypochlorite (which may damage carti- (30), respectively. Recent in vitro work planation suggested by Brody et al. is lage) from hydrogen peroxide and chlo- by Cronstein et al. supported this hypoth- that MTX could inhibit the binding of ride (55). It also inhibits leukotriene D4 esis. They showed that MTX enhanced IL-1b to the IL-1 receptor on PBMNC dipeptidase in vitro (56), but has no ef- the intracellular accumulation of 5-amin- in vitro without decreasing the produc- fect on human neutrophil lipoxygenase oimidazole-4-carboxamide ribonucle- tion of IL-1b (44). Although only 50% (57). otide and the release of adenosine from of patients receiving MTX showed a D-pen-inhibited T cell proliferation to injured neutrophils, fibroblasts, and en- sharp decrease in serum (43) and syno- mitogenic stimulation is perhaps due to dothelial cells (31); adenosine is known vial fluid IL-1b (30) in vivo, this was as- the generation of peroxide (58) or to the to inhibit leucocyte migration. Subse- sociated with a decrease in the number action of the thiol group (49), as previ- quently, using a murine carrageenan in- of painful joints. ously discussed. D-pen is known to cause flamed air pouch model, they were able MTX inhibited the spontaneous pro- a number of autoimmune conditions in- to confirm this finding in vivo (32). Fur- duction of IgM-RF and IgA-RF in vivo cluding myasthenia gravis and systemic thermore, both MTX and adenosine were (45), but clinical improvement was cor- lupus erythematosus (SLE). O’Donnell shown to inhibit leucocyte-endothelial related only with decreased IgM-RF. et al. showed in a murine model that D- cell adhesion induced by platelet-activat- This may be due to the inhibition of B pen induced the development of D-pen- ing factor. The inhibition could be par- cell proliferation and differentiation. specific T cells which were CD4+ and tially reversed by adenosine deaminase However, since improvement with MTX restricted by MHC class II molecules and the adenosine A2 receptor antago- is not limited to seropositive patients, this (59). These T cells responded to drug- nist, but not the A1 receptor antagonist. is unlikely to be the main therapeutic ef- haptenated stimulator cells but not to Furthermore, MTX reduced PMN che- fect of MTX. unstimulated cells or free drug. Interest- motaxis by suppressing leukotriene B4 ingly, generation of this D-pen-derived production, decreasing neutral protease D- antigenic determinant for T cells did not and serine protease activity (33), and in- D-penicillamine (D-Pen)has been shown require intracellular processing. This hibiting the production of superoxides to be clinically effective in RA (46); its may be the mechanism through which (34). main anti-rheumatic action is thought to D-pen induces an autoimmune disease Ex vivo MTX did not inhibit PBMNC be mediated through the action of its thiol such as myasthenia gravis. In vitro, D- proliferation in a high-folate medium, group. Since many DMARDs, including pen causes DNA breaks (60), and in but in a low-folate medium RNA pro- aurothiomalate, penicillin and levam- treated RA patients there are reported duction decreased by more than 80% isole, contain a thiol group, some authors decreases in the lymphocyte number as (35). Interestingly, Constantin et al. have have suggested that thiol itself may have well as reduced CD4:CD8 ratios. shown that the PHA-stimulated PBMNC disease-modifying properties. Thiols are Thiol-disulphide exchange inhibits the of MTX-treated RA patients have a metal chelators (47) and stabilize pro- accessory cell function of monocytes higher IL-4 gene expression than patients teins, preventing their reacting with other (51). Handel et al. showed that D-pen is with active RA (36). Recently, Genestier moieties (48). In addition, T lymphocytes an inhibitor of AP-1 binding to DNA et al. have shown that MTX at 0.1 - 10 (49), NK cells (50), and monocytes (51) through the formation of sulphur-con- mM induced an apoptosis of activated T contain cell-surface sulphydryl groups taining radicals (61). Since AP-1 is an cells (37) that was independent of Fas that are important for their function. The important transcription activator of many and CD95L ligation. thiol group can reduce disulphides by cytokine and metalloproteinase genes, Barrera et al. showed that in vivo MTX thiol-sulphide interchange, and hence this may explain the anti-inflammatory treatment is associated with decreases in can alter cell surface receptors and their properties of D-pen.

S-22 Hypothalamus-pituitary-adrenocortical and -gonadal axis in RA / M.Mechanisms Cutolo of action of second-line agents / E. ChoyEDITORIAL & G. Panayi Other in vitro effects of D-pen include of neutrophil apoptosis which was abro- logically inactive until metabolised, pri- decreased production of immunoglobu- gated by reactive oxygen species (72). marily in erythrocytes and in the liver, lin by B cells (62) and inhibition of en- Sulphapyridine inhibited NK cell activ- to 6-thioinosinic acid and 6-thioguany- dothelial-derived growth factor-stimu- ity in vitro (73), but the clinical signifi- lic acid. As an anti-metabolite, it is toxic lated neovascularisation (54). In vivo, the cance of this finding is not known. to cells in the S phase of the mitotic cy- serum concentration of IgM and, to a Danis et al. showed a decrease in in vivo cle. It acts mainly by inhibiting the func- lesser extent, IgG declines after treat- serum levels of IL-1ß, IL-6, and TNFa tion of T, B and NK cells. ment (63). following treatment with SASP. How- AZA inhibits in vitro mixed lymphocyte ever, these changes may be effects of dis- culture only when it is added during the Sulphasalazine ease improvement rather than a direct first 24 hours of culture (84). Further- Sulphasalazine (SASP) consists of sul- action of SASP (74). Bissonnette et al. more, in animal models T cell-depend- phapyridine and 5-aminosalicylate. It is showed that SASP in vitro inhibited the ent immune responses, such as delayed used extensively for the treatment of RA release of TNFa from mast cells (75), hypersensitivity and graft rejection, are and inflammatory bowel diseases. How- while Wahl et al. showed that SASP in- particularly sensitive to the action of the ever, the mechanisms of action in the two hibited the TNFa , LPS and phorbol es- drug. diseases are probably different, since 5- ter-induced expression of NF-kb (76). Abdou et al. showed that AZA inhibited aminosalicylate alone is effective in in- Since NF-kb is a major transcription fac- B cell proliferation in vitro (85) and Levy flammatory bowel disease but not in RA. tor for a number of pro-inflammatory et al. showed that the in vivo production Both sulphapyridine and 5-aminosali- cytokines, SASP could inhibit inflamma- of immunoglobulin was reduced after a cylate are known to have antiinflamma- tion through its anti-cytokine effect. few months of treatment (86). Inter- tory effects, and both are probably re- Recent evidence has suggested that an estingly, antibody responses directed quired for maximal effect in RA. important mechanism of action of SASP against thymus-dependant antigens, Some workers have suggested that the is the inhibition of leucocyte trafficking. which require the presence of T helper antibacterial effect of sulphapyridine SASP, but not sulphapyridine, inhibited cells, are more sensitive to the effect of may be important in the therapeutic ac- the activation-induced up-regulation of AZA than those directed against the thy- tion of SASP, as some antibiotics such CD11b/CD18 (MAC-1), an important mus-independent antigens. This suggests as minocycline have been shown to re- adhesion molecule, by granulocytes and that some of the suppressive effect on duce disease activity in RA (64). In vivo monocytes (77). This could potentially antibody production is mediated through SASP treatment resulted in a reduction reduce leucocyte trafficking into inflam- T cell inhibition. in Clostridium perfringens cultured from matory sites. Furthermore, similar to Cseuz et al. has shown that AZA inhib- stool samples (65). However, other MTX, SASP also increased extracellu- its the function of NK cells both in vitro bowel flora did not appear to be affected lar adenosine levels and could thereby and ex vivo (87, 88). However, treatment and these changes in bowel flora were inhibit leucocyte trafficking (78). with AZA did not produce any consist- not correlated with clinical improve- Sharon et al. showed that SASP inhib- ent fall in the NK cell numbers in vivo. ment. One interesting recent finding was ited bovine endothelial cell proliferation The decrease in NK cell function was that treatment with steroid together with in vitro (79), and this has been subse- not associated with changes in disease SASP seemed to have a deleterious ef- quently confirmed in human endotheli- activity following treatment. fect (66). The mechanism of this interac- al cells (80). Since angiogenesis is an im- Unlike MTX, AZA did not affect the se- tion is unclear, but corticosteroids may portant component of rheumatoid syno- rum level of soluble cytokine receptors, antagonise the antibacterial action of vial hypertrophy, the inhibition of en- although the serum IL-6 level was re- SASP. dothelial cell proliferation may limit dis- duced (38). The decrease in IL-6 paral- One of the most potent immunomodu- ease. 5-aminosalicylate has been shown lelled the improvement in RA disease latory effects of SASP is on B cells (67). to scavenge oxygen and hypochlorite ra- activity, although this may be a conse- SASP inhibited B cell proliferation in dicals in vitro (81). This is supported by quence of disease improvement rather vitro whilst in vivo it can lead to hypo- a clinical study from Bradley et al in 19 than a direct therapeutic action of AZA. gammaglobulinaemia (68), decreased RA patients treated with SASP. Super- antibody production to orally adminis- oxide levels fell in the responders but not Antimalarials tered antigen (69), and selective IgA in the non-responders (82). SASP inhi- Antimalarials, such as hydroxychlor- deficiency. Samanta et al. showed that bited IL-1b-induced prostaglandin E2 oquine and , have been used SASP inhibited peripheral blood mono- and glycosaminoglycan release from extensively in the treatment of SLE and nuclear cell (PBMC) proliferation to rabbit chondrocytes in vitro, suggesting RA. Their exact mechanism of action re- PHA (70) ex vivo, an effect seen only in that SASP may affect joint damage (83). mains unknown. Antimalarials accumu- patients who responded clinically to the late avidly in acid-vesicle environments drug. SASP can also inhibit proliferation such as lysosomes and the Golgi appa- of synoviocytes and reduce release of IL- Azathioprine (AZA) is an oral purine ratus (89, 90). They affect the pH of the 1b and IL-6 in a dose-dependent man- analogue which interferes with the syn- organelles, thereby inhibiting the action ner (71). SASP also increased the rate thesis of adenosine and guanine. It is bio- of certain enzymes such as acid protease,

S-23 Mechanisms of action of second-line agents / E. Choy & G. Panayi cathepsin B, and phospholipase A2 (91). unit of NF-AT, resulting in its transloca- duction in the circulating level of auto- The inhibition of such enzymes would tion from the cytoplasm into the nucleus antibodies and immune complexes (111), have various effects, depending on the to form a competent transcriptional ac- which renders this drug particularly use- cell type. Phagocytosis and the cleavage tivator (103). ful in treating the antibody-mediated sys- of peptides by enzymes such as cathep- NF-AT is an important transcription fac- temic complications of RA such as vas- sin are vital for antigen processing. Thus, tor for the production of IL-2 (104). Re- culitis. At high doses CYC also causes a the accumulation of antimalarials in the cent work has shown that CSA binds to reduction in lymphocyte numbers, espe- lysosomes will result in defective anti- immunophilins, in particular cyclophilin cially of CD8+ cells. This is due to the gen processing and presentation (89). (Cyp), which has enzymatic functions binding of CYC covalently to DNA, Antimalarials have also been shown to and regulates protein folding during pro- which prevents replication (112). The inhibit in vitro antigen processing and tein synthesis. The Cyp/CSA complex toxicity of CYC, involving late malig- presentation (90), as well as IL-1 release can bind to calcineurin and calmodulin, nancies, limits its clinical use in RA only (91). thereby inhibiting its phosphatase activ- to patients with life-threatening vasculi- In PMN, antimalarials inhibit chemo- ity. This prevents the translocation of tis and other severe extra-articular mani- taxis and phagocytosis (92, 93). In addi- transcription factor into the nucleus and festations of disease tion, they inhibit phospholipase A2 inhibits the gene expression of IL-2 and which is important in the production of other cytokines. Recently, Matsuda et al. arachidonate, the precursor of prosta- have shown that CSA can also inhibit in- Leflunomide is a isoxazol derivative and glandins. Anti-malarials bind in vitro to tracellular kinases, including MKK6 and a new immunomodulator which has been DNA, affecting DNA and RNA polymer- MMK7 (105). used in the treatment of RA. Approxi- ase as well as increasing chromosome Although CyA is primarily a T cell- mately 50% of leflunomide-treated pa- breakage (94). This may explain its in- directed drug, some evidence has been tients achieved the ACR 20% criteria of hibitory action on mitogen-induced T reported for its effects on other cell types clinical improvement in a double-blind cell proliferation both in vitro and ex vivo (106). In human monocytes and macro- placebo-controlled trial in RA, in which (95, 96). Antimalarials inhibit antibody phages, CSA induced apoptosis and it was as efficacious as SASP (113). Lef- production to the live rabies vaccine, abolished the inositol 1,4,5-triphosphate- lunomide is metabolised by the liver to though not to the killed typhoid vaccine mediated release of calcium ions from A771726, which is the active compound (97). This may be secondary to their ef- intracellular stores. CSA inhibits nitric (114). The main in vitro action of leflu- fect on antigen presentation. Antimalar- oxide synthesis in fibroblast cell lines in nomide is the inhibition of T cell prolif- ial treatment has been shown to protect vitro. This effect is independent of cal- eration stimulated by either mitogen or cartilage from damage by prostaglandins cineurin inhibition (107). In human gin- antigens (115, 116). Leflunomide inhib- in vitro (98); however, there is no clear gival fibroblasts stimulated by LPS, CSA its T cell proliferation by slowing cell clinical evidence that they retard radio- inhibited collagenase gene expression by cycling by binding to dihydro-orotate logic progression in RA. suppressing the transcription activator dehydrogenase, an enzyme involved in AP-1 (108). Low-dose CSA inhibited de novo pyrimidine synthesis (117, Cyclosporin A endothelial cell proliferation, chemotax- 118). This latter pathway is particularly Cyclosporin A (CSA) is a cyclic ende- is, and the release of metalloproteinases vital for activated rather than resting capeptide isolated from the fungi Tolypo- 2 and 9, both in vitro and in vivo (109). lymphocytes. Hence the claimed favour- clasium inflatum and Cylindrocarpon One of the most important side effects able therapeutic/toxicity window of lef- lucidium. CSA has established itself as of cyclosporin is nephrotoxicity. Recent lunomide. an important drug in transplantation and evidence suggests that this is mediated in the treatment of a number of autoim- through TGFb, as anti-TGBb antibody Combination DMARD therapy mune diseases, including RA (99). It is abrogated renal histopathologic changes It is difficult to make rational decisions a potent immunomodulator which pri- in cyclosporin-treated animals (110). regarding combination therapies for the marily inhibits T cells. In vitro CSA is treatment of RA. The evidence for the known to inhibit IL-2 secretion in vitro Cyclophosphamide modes of action of the various drugs when PBMC are stimulated either by mi- Cyclophosphamide (CYC) is a power- available is fragmentary and bedevilled togen or antigen (100,101). This effect is ful immunosuppressant widely used in by methodological problems, not the mediated through the inhibition of IL-2 the treatment of malignancy, SLE and least being the difficulty of extrapolat- gene transcription through an effect on vasculitides. In RA, it improves symp- ing from in vitro phenomena to in vivo the nuclear transcription factor of acti- toms and retards radiologic progression responses. For example, it is difficult to vated cells (NF-AT) (102). During T cell (1). CYC not only inhibits cells in the account for the fact that intramuscular activation, extra-cellular signals lead to pre-mitotic (G2) phase, but also inhibits gold salts take several weeks to have an a sharp rise in intracellular calcium. This a number of metabolic pathways. Its ma- effect, while the in vitro experiments pur- binds to calmodulin, which in turn binds jor therapeutic action in RA, however, porting to analyse their mode of action to calcineurin; the activated calcineurin is likely to be its inhibition of B cell func- are short term. Table I shows a summary dephosphorylates the cytoplasmic sub- tion. At low doses, CYC leads to a re- of the main actions of DMARDs.

S-24 Hypothalamus-pituitary-adrenocortical and -gonadal axis in RA / M.Mechanisms Cutolo of action of second-line agents / E. ChoyEDITORIAL & G. Panayi

Table I. Summary of the main actions of disease-modifying antirheumatic drugs (DMARDs). 1977; 59: 455-66. 11. LIES RB, CARDIN C, PAULUS HE: Inhibition Drug Mechanism of action by gold of human lymphocyte stimulation. An in vitro study. Ann Rheum Dis 1977; 36: 216-8. Antimalarials Inhibits antigen processing 12. SFIKAKIS PP, SOULIOTIS VL, PANAYIOTIDIS PP: Suppression of interleukin-2 and inter- Intramuscular gold salts Inhibits monocyte function leukin-2 receptor biosynthesis by gold com- Down-regulates endothelial cell selectins pounds in in vitro activated human peripheral D-penicillamine Inhibits immunocompetent cells blood mononuclear cells. Arthritis Rheum Sulphasalazine Decreases immunoglobulin production, especially IgA 1993; 36: 208-12. 13. VERWILGHEN J, KINGSLEY GH, GAMBLING Methotrexate Inhibits cell migration L, PANAYI GS: Activation of gold-reactive T Leflunomide Inhibits T cells lymphocytes in rheumatoid arthritis patients Cyclosporin A Inhibits T cells treated with gold. Arthritis Rheum 1992; 35: 1413-8. Azathioprine Inhibits T cells 14. MENARD HA, BEAUDET F, DAVIS P, HARTH Cyclophosphamide Inhibits B cells M, PERCY JS, RUSSELL AS, THOMPSON JM: Gold therapy in rheumatoid arthritis. Interim report of the Canadian multicenter prospective Thus, antimalarials and intramuscular standing of the pathogenesis of RA have trial comparing and gold should demonstrate some additive helped in the investigation of many new auranofin. J Rheumatol 1982; 8 (Suppl.): 179- effect, and this has indeed been shown possibilities. Perhaps the greatest poten- 83. to be the case (119). However, although 15. KAWAKAMI A, EGUCHI K, MIGITA K et al .: tial of research into the modes of action Inhibitory effects of gold sodium thiomalate the combination of antimalarials with D- of DMARDs is to develop better and less on the proliferation and interferon-gamma in- pen should have additive or synergistic toxic treatments directed at the same tar- duced HLA-DR expression in human endothe- effects, clinical experience is that the gets. lial cells. J Rheumatol 1990; 17: 430-5. combination is not effective and shows 16. SIN YM, WONG MK: Effect of sodium auro- thiomalate on carrageenan induced inflamma- increased toxicity (120). By contrast, References tion of the air pouch in mice. Ann Rheum Dis O’Dell and colleagues have clearly 1. IANNUZZI L, DAWSON N, ZEIN N, KUSHNER 1992; 51: 112-6. shown that triple therapy with antima- I: Does drug therapy slow radiographic deter- 17. CORKILL MM, KIRKHAM BW, HASKARD larials, SASP and MTX is effective and ioration in rheumatoid arthritis ? N Engl J Med DO, BARBATIS C, GIBSON T, PANAYI GS: 1983; 309: 1023-8. Gold treatment of rheumatoid arthritis de- without increased toxicity (121). Could 2. PANAYI GS, LANCHBURY JS, KINGSLEY GH: creases synovial expression of the endothelial this outcome have been predicted from The importance of the T cell in initiating and leukocyte adhesion receptor ELAM-1. J Rheu- maintaining the chronic synovitis of rheuma- matol 1991; 18: 1453-60. the known properties of the drugs ? The antimalarial, by inhibiting antigen pro- toid arthritis. Arthritis Rheum 1992; 35: 729- 18. ROSENBERG SA, LIPSKY PE: Inhibition of 35. pokeweed mitogen-induced immunoglobulin cessing, would have inhibited T cell ac- 3. FELDMANN M, BRENNAN FM, MAINI RN: production in humans by gold compounds. J tivation, while the MTX would have in- Rheumatoid arthritis. Cell 1996; 85: 307-10. Rheumatol 1979; 5 (Suppl.): 107-11. hibited cytokine release and cell migra- 4. FORESTIER J: Rheumatoid arthritis and its 19. OLSEN N, ZIFF M, JASIN HE: Spontaneous tion into the synovium. The SASP would treatment by gold salts. The results of six years’ synthesis of IgM rheumatoid factor by blood experience. J Clin Lab Med 1935; 20: 827-40. mononuclear cells from patients with rheuma- have inhibited immunoglobulin produc- 5. EMPIRE RHEUMATISM COUNCIL RS: Gold toid arthritis: Effect of treatment with gold salts tion and, presumably, the generation of therapy in rheumatoid arthritis. Report of a or D-penicillamine. J Rheumatol 1984; 11: 17- immune complexes. An animal model in multi-centre controlled trial. Ann Rheum Dis 21. which these combinations could first be 1960; 19: 95-119. 20. HIGHTON J, PANAYI GS, SHEPHERD P, FAITH 6. RUDKOWSKI R, ZIEGLER JB, GRAHAM GG, A, GRIFFIN J, GIBSON TG: Fall in immune tried out is clearly needed. JOULIANOS G: Gold complexes and activation complex levels during gold treatment of rheu- With the rapid growth of molecular ge- of human polymorphonuclear leukocytes. Dis- matoid arthritis. Ann Rheum Dis 1981; 40: 575- netics and, in particular, pharmacogen- sociation of changes in membrane potential 9. etics, increasing knowledge will remove and oxidative burst. Biochem Pharmacol 1992; 21. OSTUNI PA, SIMIONI M, MARSON P, TRAVA- 44: 1091-8. GLIA P, VOLANTE D, GAMBARI PF: Serum much of the guesswork from this predic- 7. FELSON DT, ANDERSON JJ, MEENAN RF: The IgA and gold toxicity in rheumatoid arthritis: tive process. Thus SASP has been shown comparative efficacy and toxicity of second- Lack of predicting value. Clin Exp Rheumatol recently to inhibit methyl- line drugs in rheumatoid arthritis. Results of 1986; 4: 359-62. transferase, which catabolises the s-me- two metaanalyses. Arthritis Rheum 1990; 33: 22. MADHOK R, CRILLY A, MURPHY E, SMITH J, 1449-61. WATSON J, CAPELL HA: Gold therapy lowers thylation of (122). Therefore, 8. FARAHAT MN, YANNI G, POSTON R, PANAYI serum interleukin 6 levels in rheumatoid ar- SASP may interfere with the metabolism GS: Cytokine expression in synovial mem- thritis. J Rheumatol 1993; 20: 630-3. of AZA. Combination therapy of SASP branes of patients with rheumatoid arthritis and 23. HARTH M, COUSIN K, MCCAIN GA: In vitro with AZA should therefore be approach- osteoarthritis. Ann Rheum Dis 1993; 52: 870- effects of two gold compounds, and D-peni- 5. cillamine on the production of interferon gam- ed with caution. 9. HARTH M, STILLER CR, ST. C, SINCLAIR NR, ma. Immunopharmacol Immunotoxicol 1990; EVANS J, MCGIRR D, ZUBERI R: Effects of a 12: 39-60. Conclusion gold salt on lymphocyte responses. Clin Exp 24. DANIS VA, KULESZ AJ, NELSON DS, BROOKS Although much uncertainty remains con- Immunol 1977; 27: 357-64. PM: The effect of gold sodium thiomalate and 10. LIPSKY PE, ZIFF MA: Inhibition of mitogen auranofin on lipopolysaccharide-induced inter- cerning the mechanism of action of and antigen-induced human lymphocyte pro- leukin-1 production by blood monocytes in DMARDs, recent advances in the under- liferation by gold compounds. J Clin Invest vitro: Variation in healthy subjects and patients

S-25 Mechanisms of action of second-line agents / E. Choy & G. Panayi

with arthritis. Clin Exp Immunol 1990; 79: 335- ripheral T cells. J Clin Invest 1998; 102: 322- cillamine and copper salts. J Immunol 1978; 40. 8. 120: 1006-13. 25. HANDEL ML, SIVERTSEN S, WATTS CK, DAY 38. BARRERA P, BOERBOOMS AM, JANSSEN EM 53. MATSUBARA T, HIROHATA K: Suppression of RO, SUTHERLAND RL : Comparative effects et al.: Circulating soluble tumor necrosis fac- human fibroblast proliferation by D-penicilla- of gold on the interactions of transcription tor receptors, interleukin-2 receptors, tumor mine and copper sulfate in vitro. Arthritis factors with DNA. Agents Actions 1993; 44 necrosis factor alpha, and interleukin-6 levels Rheum 1988; 31: 964-72. (Suppl.): 219-23. in rheumatoid arthritis. Longitudinal evalua- 54. MATSUBARA T, SAURA R, HIROHATA K, ZIFF 26. GOLDBERG RL, KAPLAN SR, FULLER GC: Ef- tion during methotrexate and azathioprine ther- MA: Inhibition of human endothelial cell pro- fect of heavy metals on human rheumatoid apy. Arthritis Rheum 1993; 36: 1070-9. liferation in vitro and neovasculisation in vivo synovial cell proliferation and collagen syn- 39. NEURATH MF, HILDNER K, BECKER C et al.: by d-penicillamine. J Clin Invest 1989; 83: thesis. Biochem Pharmacol 1983; 32: 2763-6. Methotrexate specifically modulates cytokine 158-67. 27. HAURANI FI, HERMAN GM, ABBOUD EM: production by T cells and macrophages in mur- 55. CUPERUS RA, HOOGLAND H, WEVER R, Thymidylate synthetase of human lympho- ine collagen-induced arthritis (CIA): A mech- MUIJSERS AO: The effect of D-penicillamine cytes augmented in vitro by methotrexate. anism for methotrexate-mediated immuno- on myeloperoxidase: Formation of compound Cancer Biochemistry Biophysics 1985; 8: 29- suppression. Clin Exp Immunol 1999; 115: 42- III and inhibition of the chlorinating activity. 33. 55. Biochim Biophys Acta 1987; 912: 124-31. 28. ALLEGRA CJ, DRAKE JC, JOLIVET J, CHAB- 40. MEYER FA, YARON I, MASHIAH V, YARON 56. HUBER M, KEPPLER D: Inhibition of leuko- NER BA: Inhibition of phosphoribosylamino- M: Methotrexate inhibits proliferation but not triene D4 catabolism by D-penicillamine. Eur imidazolecarboxamide transformylase by interleukin 1 stimulated secretory activities of J Biochem 1987; 167: 73-9. methotrexate and dihydrofolic acid polyglu- cultured human synovial fibroblasts. J Rheu- 57. NIELSEN OH, AHNFELT-RONNE I, ELM- tamates. Proc Natl Acad Sci USA 1985; 82: matol 1993; 20: 238-42. GREEN J: A comparison of the effect of time- 4881-5. 41. SEGAL R, MOZES E, YARON M, TARTAKOV- gadine, levamisole, and D-penicillamine on hu- 29. VAN DE KERKHOF PC, BAUER FW, MAASSEN SKY B: The effects of methotrexate on the pro- man neutrophil metabolism of endogenous ara- DE, GROOD RM: Methotrexate inhibits the duction and activity of interleukin-1. Arthritis chidonic acid and chemotaxis. Pharmacol leukotriene B4 induced intraepidermal accu- Rheum 1989; 32: 370-7. Toxicol 1988; 62: 322-5. mulation of polymorphonuclear leukocytes. Br 42. SEITZ M, ZWICKER M, LOETSCHER P: Effects 58. MITA S, MATSUNAGA K: Differences in the J Dermatol 1985; 113: 251a-5a. of methotrexate on differentiation of mono- effects of the antirheumatic drugs, bucillamine 30. THOMAS R, CARROLL GJ: Reduction of leu- cytes and production of cytokine inhibitors by and D-penicillamine, on mitogen-induced pro- kocyte and interleukin-1 beta concentrations monocytes. Arthritis Rheum 1998; 41: 2032- liferation of mouse spleen cells. Agents Actions in the synovial fluid of rheumatoid arthritis pa- 8. 1990; 30: 363-8. tients treated with methotrexate. Arthritis 43. CHANG DM, WEINBLATT ME, SCHUR PH: 59. O’DONNELL CA, COLEMAN JW: A T-cell re- Rheum 1993; 36: 1244-52. The effects of methotrexate on interleukin 1 sponse to the anti-arthritic drug penicillamine 31. CRONSTEIN BN, EBERLE MA, GRUBER HE, in patients with rheumatoid arthritis. J Rheum- in the mouse: requirements for generation of LEVIN RI: Methotrexate inhibits neutrophil atol 1992; 19: 1678-82. the drug-derived antigen. Immunology 1992; function by stimulating adenosine release from 44. BRODY M, BOHM I, BAUER R: Mechanism of 76: 604-9. connective tissue cells. Proc Natl Acad Sci USA action of methotrexate: Experimental evidence 60. YAMANAKA H, HAKODA M, KAMATANI N, 1991; 88: 2441-5. that methotrexate blocks the binding of inter- KASHIWAZAKI S, CARSON DA: Formation of 32. CRONSTEIN BN, NAIME D, OSTAD E: The leukin 1 beta to the interleukin 1 receptor on DNA strand breaks by D-penicillamine and antiinflammatory mechanism of methotrexate. target cells. Eur J Clin Chem Clin Biochem bucillamine in human lymphocytes. Immuno- Increased adenosine release at inflamed sites 1993; 31: 667-74. pharmacology 1993; 26: 113-8. diminishes leukocyte accumulation in an in 45. OLSEN NJ, CALLAHAN LF, PINCUS T: Immu- 61. HANDEL ML, WATTS CK, SIVERTSEN S, DAY vivo model of inflammation. J Clin Invest nologic studies of rheumatoid arthritis patients RO, SUTHERLAND RL: D-penicillamine caus- 1993; 92: 2675-82. treated with methotrexate. Arthritis Rheum es free radical-dependent inactivation of acti- 33. SPERLING RI, BENINCASO AI, ANDERSON 1987; 30: 481-8. vator protein-1 DNA binding. Mol Pharmacol RJ, COBLYN JS, AUSTEN KF, WEINBLATT 46. ANONYMOUS: Controlled trial of D(-)peni- 1996; 50: 501-5. ME: Acute and chronic suppression of leuko- cillamine in severe rheumatoid arthritis. Lan- 62. BAIER-BITTERLICH G, SCHUMACHER W, triene B4 synthesis ex vivo in neutrophils from cet 1973; 1: 275-80. SCHULZ, TF, WACHTER H, DIERICH MP: Ef- patients with rheumatoid arthritis beginning 47. BACON PA, SALMON M : Modes of action of fects of D-penicillamine on human cell lines. treatment with methotrexate. Arthritis Rheum second-line agents. Scand J Rheumatol 1987; Arzneimittel-Forschung 1993; 43: 395-8. 1992; 35: 376-84. 64 (Suppl.): 17-24. 63. WERNICK R, MERRYMAN P, JAFFE I, ZIFF M: 34. AL BALLA S, JOHNSTON C, DAVIS P: The in 48. GERBER DA: Copper-catalyzed thermal aggre- IgG and IgM rheumatoid factors in rheuma- vivo effect of nonsteroidal anti-inflammatory gation of human gamma-globulin. Inhibition toid arthritis. Quantitative response to peni- drugs, gold sodium thiomalate and methotrex- by histidine, gold thiomalate, and penicilla- cillamine therapy and relationship to disease ate on neutrophil superoxide radical genera- mine. Arthritis Rheum 1974; 17: 85-91. activity. Arthritis Rheum 1983; 26: 593-8. tion. Clin Exp Rheumatol 1990; 8: 41-5. 49. THORNE KJ, FREE J, FRANKS D: Role of sul- 64. KLOPPENBURG M, BREEDVELD FC, MILTEN- 35. OLSEN NJ, MURRAY LM: Antiproliferative ef- phydryl groups in T lymphocyte-mediated cy- BURG AM, DIJKMANS BA: Antibiotics as dis- fects of methotrexate on peripheral blood mo- totoxicity. Clin Exp Immunol 1982; 50: 644- ease modifiers in arthritis. Clin Exp Rheumatol nonuclear cells. Arthritis Rheum 1989; 32: 378- 50. 1993; 11 (Suppl. 8): S113-5. 85. 50. RISTOW SS, STARKEY JR, STANFORD DR, 65. BRADLEY SM, NEUMANN VC, BARR K et al.: 36. CONSTANTIN A, LOUBET-LESCOULIE P, DAVIS WC, BROOKS CG: Cell surface thiols, Sequential study of bacterial antibody levels LAMBERT N et al.: Antiinflammatory and im- but not intracellular glutathione, are essential and faecal flora in rheumatoid arthritis patients munoregulatory action of methotrexate in the for cytolysis by a cloned murine natural killer taking sulphasalazine. Br J Rheumatol 1993; treatment of rheumatoid arthritis: Evidence of cell line. Immunol Invest 1985; 14: 401-14. 32: 683-8. increased interleukin-4 and interleukin-10 gene 51. MCKEOWN MJ, HALL ND, CORVALAN JR: 66. GOUGH A, SHEERAN T, ARTHUR V, PANAYI expression demonstrated in vitro by competi- Defective monocyte accessory function due to G, EMERY P: Adverse interaction between in- tive reverse transcriptase-polymerase chain re- surface sulphydryl (SH) oxidation in rheuma- tramuscular methylprednisolone and sulpha- action. Arthritis Rheum 1998; 41: 48-57. toid arthritis. Clin Exp Immunol 1984; 56: 607- salazine in patients with early rheumatoid ar- 37. GENESTIER L, PAILLOT R, FOURNEL S, FER- 13. thritis. A pilot study. Scand J Rheumatol 1994; RARO C, MIOSSEC P, REVILLARD JP: Immu- 52. LIPSKY PE, ZIFF M: The effect of D-penicilla- 23: 46-8. nosuppressive properties of methotrexate: mine on mitogen-induced human lymphocyte 67. IMAI F, SUZUKI T, ISHIBASHI T, DOHI Y: Ef- Apoptosis and clonal deletion of activated pe- proliferation: synergistic inhibition by D-peni- fect of on B cells. Clin Exp Rheu-

S-26 Hypothalamus-pituitary-adrenocortical and -gonadal axis in RA / M.Mechanisms Cutolo of action of second-line agents / E. ChoyEDITORIAL & G. Panayi

matol 1991; 9: 259-64. 82. BRADLEY SM, LE GALLEZ P, THROUGHTON 99. PANAYI GS, TUGWELL P: The use of cyclo- 68. FARR M, KITAS GD, TUNN EJ, BACON PA: Im- PR, GOOI HC, ASTBURY C, BIRD HA: The ef- sporin A in rheumatoid arthritis: Proceedings munodeficiencies associated with sulphasal- fect of sulphasalazine on neutrophil superoxide of an international consensus meeting. Br J azine therapy in inflammatory arthritis. Br J generation in rheumatoid arthritis. Br J Rheum- Rheumatol 1993; 32 (Suppl. 1): 1-3. Rheumatol 1991; 30: 413-7. atol 1997; 36: 530-4. 100. KRONKE M, LEONARD WJ, DEPPER JM et al.: 69. SHELDON P, PELL P: Comparison of the effects 83. NOSE M, SASANO M, KAWASHIMA Y: Sala- Cyclosporin A inhibits T-cell growth factor of sulphasalazine, 5-aminosalicylic acid and zosulfapyridine suppresses chondrocyte medi- gene expression at the level of mRNA tran- sulphapyridine on the humoral response to an- ated degradation induced by interleukin 1beta. scription. Proc Natl Acad Sci USA 1984; 81: tigen in vivo. Adv Exp Med Biol 1995; 371B: J Rheumatol 1997; 24: 550-4. 5214-8. 905-8. 84. AL-SAFI SA, MADDOCKS JL: Effects of aza- 101. GRANELLI PIPERNO A, ANDRUS L, STEIN- 70. SAMANTA A, WEBB C, GRINDULIS KA, FLE- thioprine on the human mixed lymphocyte re- MAN RM: Lymphokine and non-lymphokine MING J, SHELDON PJ: Sulphasalazine therapy action (MLR). Br J Clin Pharmacol 1983; 15: mRNA levels in stimulated human T cells. Ki- in rheumatoid arthritis: Qualitative changes in 203-9. netics, mitogen requirements, and effects of lymphocytes and correlation with clinical re- 85. ABDOU NI, ZWEIMAN B, CASELLA SR: Effects cyclosporin. J Exp Med 1986; 163: 922-37. sponse. Br J Rheumatol 1992; 31: 259-63. of azathioprine therapy on bone marrow-de- 102. WIEDERRECHT G, LAM E, HUNG S, MARTIN 71. AONO H, HASUNUMA T, FUJISAWA K, pendent and thymus-dependent cells in man. M, SIGAL N: The mechanism of action of FK- NAKAJIMA T, YAMAMOTO K, MITA S, Clin Exp Immuno 1973; 13: 55-64. 506 and cyclosporin A. Ann NY Acad Sci NISHIOKA K: Direct suppression of human 86. LEVY J, BARNETT EV, MACDONALD NS, 1993; 696: 9-19. synovial cell proliferation in vitro by salazo- KLINENBERG JR, PEARSON CM: The effect of 103. BRAM RJ, HUNG DT, MARTIN PK, SCHREIB- sulfapyridine and bucillamine. J Rheumatol azathioprine on gammaglobulin synthesis in ER SL, CRABTREE GR: Identification of the 1996; 23: 65-70. man. J Clin Invest 1972; 51: 2233-8. immunophilins capable of mediating inhibi- 72. AKAHOSHI T, NAMAI R, SEKIYAMA N, 87. CSEUZ R, PANAYI GS: The inhibition of NK tion of signal transduction by cyclosporin A TANAKA S, HOSAKA S, KONDO H: Rapid in- cell function by azathioprine during the treat- and FK506: Roles of calcineurin binding and duction of neutrophil apoptosis by sulfasal- ment of patients with rheumatoid arthritis. Br cellular location. Mol Cell Biol 1993; 13: azine: implications of reactive oxygen species J Rheumatol 1990; 29: 358-62. 4760-9. in the apoptotic process. J Leukocyte Biol 1997; 88. CSEUZ R, BARNES P, PANAYI GS: Natural 104. MCCAFFREY PG, PERRINO BA, SODERLING 62: 817-26. killer cells in the blood of patients with rheu- TR, RAO A: NF-ATp, a T lymphocyte DNA- 73. GIBSON PR, JEWELL DP: Sulphasalazine and matoid arthritis treated with azathioprine. Br binding protein that is a target for calcineurin derivatives, natural killer activity and ulcera- J Rheumatol 1990; 29: 284-7. and immunosuppressive drugs. J Biol Chem tive colitis. Clin Sci 1985; 69: 177-84. 89. KROGSTAD DJ, SCHLESINGER PH: Acid- 1993; 268: 3747-52. 74. DANIS VA, FRANIC GM, RATHJEN DA, vesicle function, intracellular pathogens, and 105. MATSUDA S, MORIGUCHI T, KOYASU S, LAURENT RM, BROOKS PM: Circulating cy- the action of chloroquine against Plasmodium NISHIDA E: T lymphocyte activation signals tokine levels in patients with rheumatoid ar- falciparum. [1]. N Engl J Med 1987; 317: 542- for interleukin-2 production involve activa- thritis: Results of a double blind trial with sul- 9. tion of MKK6-p38 and MKK7-SAPK/JNK phasalazine. Ann Rheum Dis 1992; 51: 946- 90. MACINTYRE AC, CUTLER DJ: Role of lyso- signaling pathways sensitive to cyclosporin 50. somes in hepatic accumulation of chloroquine. A. J Biol Chem 1998; 273: 12378-82. 75. BISSONNETTE EY, ENCISO JA, BEFUS AD: In- J Pharm Sci 1988; 77: 196-9. 106. YOCUM D: Immunological actions of cyclo- hibitory effects of sulfasalazine and its meta- 91. GINSBURG H, GEARY TG: Current concepts sporin A in rheumatoid arthritis. Br J Rheum- bolites on histamine release and TNF-alpha and new ideas on the mechanism of action of atol 1993; 32 (Suppl. 1): 38-41. production by mast cells. J Immunol 1996; 156: -containing antimalarials. [1]. Bio- 107. TRAJKOVIC V, BADOVINAC V, JANKOVIC 218-23. chem Pharmacol 1987; 36: 1567-76. V, SAMARDZIC T, MAKSIMOVIC D, POPA- 76. WAHL C, LIPTAY S, ADLER G, SCHMID RM: 92. WARD PA: The chemosuppression of chemo- DIC D: Cyclosporin A suppresses the induc- Sulfasalazine: A potent and specific inhibitor taxis. J Exp Med 1966; 124: 209-26. tion of nitric oxide synthesis in interferon- of nuclear factor kappa B. J Clin Invest 1998; 93. RHODES JM, MCLAUGHLIN JE, BROWN DJ, gamma-treated L929 fibroblasts. Scand J 101: 1163-74. NUTTALL LA, JEWELL DP: Inhibition of leu- Immunol 1999; 49: 126-30. 77. GREENFIELD SM, HAMBLIN AS, SHAKOOR cocyte motility and prevention of immune- 108. SUGANO N, ITO K, MURAI S: Cyclosporin A ZS, TEARE JP, PUNCHARD NA, THOMPSON complex experimental colitis by hydroxychlor- inhibits collagenase gene expression via AP- RP: Inhibition of leucocyte adhesion molecule oquine. Gut 1982; 23: 181-7. 1 and JNK suppression in human gingival upregulation by tumor necrosis factor alpha: 94. PANAYI GS, NEILL WA, DUTHIE JJR, fibroblasts. J Periodont Res 1998; 33:448-52. A novel mechanism of action of sulphasala- MCCORMICK JN: The action of chloroquine 109. IURLARO M, VACCA A, MINISCHETTI M et zine. Gut 1993; 34: 252-6. phosphate in rheumatoid arthritis. 11 Chromo- al.: Antiangiogenesis by cyclosporine. Exp 78. GADANGI P, LONGAKER M, NAIME D et al.: some damaging effect. Ann Rheum Dis 1973; Hematol 1998; 26: 1215-22. The anti-inflammatory mechanism of sulfa- 32: 547. 110. KHANNA AK, CAIRNS VR, BECKER CG, salazine is related to adenosine release at in- 95. TRIST DG, WEATHERALL M: Inhibition of HOSENPUD JD: Transforming growth factor flamed sites. J Immunol 1996; 156: 1937-41. lymphocyte transformation by mepacrine and (TGF)-beta mimics and anti-TGF-beta anti- 79. SHARON P, DRAB EA, LINDER JS, WEIDMAN chloroquine. J Pharm Pharmacol 1981; 33: body abrogates the in vivo effects of cyclo- SW, SABESIN SM, RUBIN DB: The effect of 434-8. sporine: Demonstration of a direct role of sulfasalazine on bovine endothelial cell pro- 96. PANAYI GS, NEILL WA, DUTHIE JJR, TGF-beta in immunosuppression and neph- liferation and cell cycle phase distribution. MCCORMICK JN: Action of chloroquine phos- rotoxicity of cyclosporine. Transplantation Comparison with olsalazine, 5-aminosalicylic phate in rheumatoid arthritis. 1. Immunosup- 1999; 67: 882-9. acid, and sulfapyridine. J Lab Clin Med 1992; pressive effect. Ann Rheum Dis 1973; 32: 316- 111. SCOTT DG, BACON PA, ALLEN C, ELSON 119: 99-107. 8. CJ, WALLINGTON T: IgG rheumatoid factor, 80. MADHOK R, WIJELATH E, SMITH J, WATSON 97. THOMPSON GR, BARTHOLOMEW L: The ef- complement and immune complexes in rheu- J, STURROCK RD, CAPELL HA: Is the benefi- fect of chloroquine on antibody production. matoid synovitis and vasculitis: Comparative cial effect of sulfasalazine due to inhibition of University of Michigan Medical Center Jour- and serial studies during cytotoxic therapy. synovial neovascularization ? J Rheumatol nal 1964; 30: 227-30. Clin Exp Immunol 1981; 43: 54-63. 1991; 18: 199-202. 98. MANKU MS, HORROBIN DF: Chloroquine, 112. BAST RC, BAST BS: Critical review of pre- 81. WILLIAMS JG, HALLETT MB: Effect of sul- quinine, procaine, quinidine, tricyclic antide- viously reported animal studies of tumor im- phasalazine and its active metabolite, 5-amino- pressants, and methylxanthines as prostaglan- munotherapy with nonspecific immunosti- salicylic acid, on toxic oxygen metabolite pro- din agonists and antagonists. Lancet 1976; 2: mulants. [2]. Ann NY Acad Sci 1976; 277: 60- duction by neutrophils. Gut 1989; 30: 1581-7. 1115-7. 93.

S-27 Mechanisms of action of second-line agents / E. Choy & G. Panayi

113. SMOLEN JS, KALDEN JR, SCOTT DL et al.: proliferation. Transplantation 1993; 55: 1361- bination therapy with gold and hydroxychlor- Efficacy and safety of leflunomide compared 6. oquine in rheumatoid arthritis: A prospective, with placebo and sulphasalazine in active 116. BARTLETT RR: Immunopharmacological randomized, placebo-controlled study. Br J rheumatoid arthritis: A double-blind, random- profile of HWA 486, a novel isoxazol deriva- Rheumatol 1989; 28: 128-33. ised, multicentre trial. European Leflunomide tive. II. In vivo immunomodulating effects dif- 120. GIBSON TG, EMERY P, ARMSTRONG RD, Study Group. Lancet 1999; 353: 259-66. fer from those of cyclophosphamide, pred- CRISP AJ, PANAYI GS: Combined D-penicil- 114. MIRMOHAMMADSADEGH A, HOMEY B, nisolone, or cyclosporin A. Int J Immuno- lamine and chloroquine treatment - A com- ABTS HF, KOHRER K, RUZICKA T, MICHEL pharmacol 1986; 8: 199-204. parative study. Br J Rheumatol 1987; 26: 279- G: Differential modulation of pro- and anti- 117. CHERWINSKI HM, COHN RG, CHEUNG P et 84. inflammatory cytokine receptors by N-(4- al.: The immunosuppressant leflunomide in- 121. O’DELL JR, HAIRE CE, ERIKSON N et al.: trifluoromethylphenyl)-2-cyano-3-hydroxy- hibits lymphocyte proliferation by inhibiting Treatment of rheumatoid arthritis with meth- crotonic acid amide (A77 1726), the physi- pyrimidine biosynthesis. J Pharmacol Exp otrexate alone, sulfasalazine and hydroxy- ologically active metabolite of the novel im- Ther 1995; 275: 1043-9. chloroquine, or a combination of all three me- munomodulator leflunomide. Biochem Phar- 118. BRUNEAU JM, YEA CM, SPINELLA-JAEGLE dications. N Engl J Med 1996; 334: 1287-91. macol 1998; 55: 1523-9. S et al.: Purification of human dihydro-orotate 122. SZUMLANSKI CL, WEINSHILBOUM RM : 115. CHONG AS, FINNEGAN A, JIANG X, GEBEL dehydrogenase and its inhibition by A77 1726, Sulphasalazine inhibition of thiopurine meth- H, SANKARY HN, FOSTER P, WILLIAMS JW: the active metabolite of leflunomide. Biochem yltransferase: A possible mechanism for in- Leflunomide, a novel immunosuppressive J 1998; 336: 299-303. teraction with 6-mercaptopurine and azathio- agent. The mechanism of inhibition of T cell 119. SCOTT DL, DAWES PT, TUNN E et al.: Com- prine. Br J Clin Pharmacol 1995; 39: 456-9.

S-28